chore: add void attention footer links 342d5f8
Taylor commited on
Spaces:
Sleeping
Sleeping
Commit History
chore: cross-link all nine spaces in footer 6553234
Taylor commited on
perf: cannon-rotated WASM SIMD kernels (14KB→22KB) 8f4ed48
Taylor commited on
perf: raise token limits to 256 (both PyTorch and Aether) ec694f7
Taylor commited on
feat: Aether v2 with RoPE fix -- PyTorch vs Aether side by side fcac5c7
Taylor commited on
perf: run both models in parallel with ThreadPoolExecutor 64ff7cb
Taylor commited on
revert: back to working PyTorch for both models 0c8d249
Taylor commited on
fix: remove missing WASM flashAttention + stream results independently e5e1d2b
Taylor commited on
fix: handle WASM OOM for LM head + show error details 90bf42d
Taylor commited on
fix: use Q8_0 GGUF + type-aware dequantization 5b5a680
Taylor commited on
perf: add WASM SIMD kernels + use Q4_K_M for faster inference 7336fde
Taylor commited on
feat: PyTorch vs Aether side-by-side inference c92238b
Taylor commited on
fix: make prompt input interactive in Gradio 5 a06fe42
Taylor commited on
fix: upgrade to Gradio 5 -- Gradio 4 incompatible with latest huggingface_hub bad02a2
Taylor commited on
fix: switch to transformers+torch(cpu) -- no C++ compilation needed 33ad9ed
Taylor commited on
fix: use pre-built llama-cpp-python wheel to avoid build timeout 53ff270
Taylor commited on
fix: switch to Gradio SDK, drop Docker build 854ee42
Taylor commited on
Upload folder using huggingface_hub 868752a verified
Upload folder using huggingface_hub b4d8480 verified
feat: add personality models tab and all models catalog 2ba37ce
Taylor commited on
fix: switch from llama-cpp-python to transformers+peft 26dd3c3
Taylor commited on
fix: switch from Docker to Gradio SDK to fix build timeout 103b9da
Taylor commited on